The energy value of pig diets estimated in vitro and in vivo

The content of digestible (DE) and metabolizable (ME) energy was determined in 29 pig diets by in vivo and in vitro procedures. Growing, castrated males were used for the in vivo balance study. The DE and ME values of diets estimated in vivo and in vitro were similar. The correlation coefficients for DE and ME were 0.802 and 0.772 , respectively. Predicting the energy value of pig diets with satisfactory precision seems possible using the in vitro technique.


INTRODUCTION
Information on the energy values of feedstuffs and diets is needed for efficient diet formulation and prediction of animal performance.The energy value of individual feedstuffs and complete diets can be expressed as the content of digestible (DE), metabolizable (ME) or net (NE) energy.The present feed evaluation system is based on analysed content and determined or tabulated values of nutrient digestibility.
A conventional pig balance study is time consuming and costly.In recent years several different in vitro methods have been developed for evaluating energy digestibility in pigs (van der Meer and Perez, 1992;Boisen and Fernández, 1997).Validation of the methods has been attempted by determining the relationship between in vitro and in vivo results (e.g., Graham et al., 1989;Boisen and Fernández, 1997).
The aim of this study was to compare the energy values (DE and ME) of pig diets estimated in balance experiments and using an in vitro method, and to assess the validity of the in vitro assay.

Methods
The experiment on pigs was carried out on growing castrated males of a body weight range from 20 to 40 kg.The animals were fed equal meals three times daily.After four days of feeding the experimental diets, faeces were collected quantitatively for 6 days.Six pigs were used per diet.The digestible energy content was calculated from the difference between energy intake and energy excreted in faeces.The metabolizable energy content was calculated from DE after correction for protein content in urine (Noblet et al., 1989).
The in vitro method developed by Boisen and Fernández (1997) was used for prediction of total digestibility of energy.Total tract digestion was simulated by three consecutive incubations corresponding to digestion in the stomach, the small intestine and the hindgut: with pepsin at pH 2.0 for 2 h, pancreatin at pH 6.8 for 4 h and a multi-enzyme complex (contain a wide range of microbial carbohydrases including arabinase, cellulase, ß-glucanase, hemicellulase, xylanase and pectinase) at pH 4.8 for 18 h at 39ºC.Digestibility of energy was calculated from the digestibility of organic matter using a regression equation.The content of DE and ME was calculated using regression equations.
Dry matter, nitrogen, ether extract, and crude fibre of diets were estimated using standard methods (AOAC, 1990).The gross energy of diets and faeces was determined by a Parr adiabatic oxygen bomb KL-10 caolorimeter.

RESULTS AND DISCUSSION
The chemical composition and energy value of 29 diets is shown in Table 1.The crude protein and ether extract contents were more variable than the content of ash, crude fibre and GE.
In general, the DE and ME contents in diets estimated in pigs and in vitro were similar (Table 1).The relationship between the DE and ME contents in diets estimated in pigs and in vitro is given in Figures 1 and 2. The correlation coefficients were high (r 2 =0.802 and 0.772, respectively).This is in agreement with results obtained by Boisen and Fernández (1997) who found a high correlation between total energy digestibility in pigs and organic matter digestibility in vitro in 31 feedstuffs (r 2 =0.94) and between total energy digestibility in pigs and predicted in vitro energy digestibility in 34 diets (r 2 =0.87).A close relationship (r 2 =0.84) between organic matter digestibility determined in pigs and in vitro using a similar in vitro method was also found by van der Meer and Perez (1992) in 89 diets.

CONCLUSIONS
It seems possible to predict the energy value of pig diets with satisfactory precision using an in vitro technique.

Figure 2 .
Figure 2. Relationship between the metabolizable energy content of 29 diets estimated on pigs (ME pig ) and in vitro (ME in vitro )

Table 1 .
Chemical composition (g kg -1 DM) and energy value (cal g -1 DM) of diets for pigs